| Metric | cnn | naive | xgboost | lstm |
|---|---|---|---|---|
| diff_of_means | 0.234 | -0.509 | -0.528 | -0.621 |
| ratio_of_sd | 0.975 | 0.997 | 0.947 | 0.926 |
| amplitude_ratio_of_means | 0.952 | 0.956 | 0.818 | 0.719 |
| maximum_error | 0.312 | 0.126 | 0.256 | 0.269 |
| ks_mean_on_coarse_res_with_extremes | 0.146 | 0.137 | 0.165 | 0.281 |
| qqplot_mae | 0.682 | 1.482 | 1.541 | 1.810 |
| acf_mae | 0.036 | 0.034 | 0.080 | 0.112 |
| extremogram_mae | 0.034 | 0.041 | 0.019 | 0.040 |
Important: Right now we are only estimating the upper tail
extremogram. Currently we didn’t find a way to estimate the two tales at
the same time. We are using quant = .97
| Metric | cnn | naive | xgboost | lstm |
|---|---|---|---|---|
| diff_of_means | 0.241 | -0.567 | -0.594 | -0.677 |
| ratio_of_sd | 0.971 | 0.967 | 0.917 | 0.895 |
| amplitude_ratio_of_means | 0.949 | 0.959 | 0.809 | 0.708 |
| maximum_error | 0.324 | 0.127 | 0.247 | 0.273 |
| ks_mean_on_coarse_res_with_extremes | 0.085 | 0.041 | 0.118 | 0.224 |
| qqplot_mae | 0.703 | 1.651 | 1.735 | 1.970 |
| acf_mae | 0.034 | 0.019 | 0.072 | 0.105 |
| extremogram_mae | 0.016 | 0.017 | 0.017 | 0.021 |
Important: Right now we are only estimating the upper tail
extremogram. Currently we didn’t find a way to estimate the two tales at
the same time. We are using quant = .97
| Metric | cnn | naive | xgboost | lstm |
|---|---|---|---|---|
| diff_of_means | 0.313 | -0.444 | -0.472 | -0.559 |
| ratio_of_sd | 0.968 | 0.995 | 0.947 | 0.923 |
| amplitude_ratio_of_means | 0.947 | 0.958 | 0.811 | 0.710 |
| maximum_error | 0.321 | 0.111 | 0.249 | 0.260 |
| ks_mean_on_coarse_res_with_extremes | 0.089 | 0.053 | 0.089 | 0.190 |
| qqplot_mae | 0.911 | 1.294 | 1.379 | 1.628 |
| acf_mae | 0.033 | 0.031 | 0.082 | 0.113 |
| extremogram_mae | 0.020 | 0.016 | 0.012 | 0.031 |
Important: Right now we are only estimating the upper tail
extremogram. Currently we didn’t find a way to estimate the two tales at
the same time. We are using quant = .97
| Metric | cnn | naive | xgboost | lstm |
|---|---|---|---|---|
| diff_of_means | 0.265 | -0.266 | -0.284 | -0.392 |
| ratio_of_sd | 0.982 | 1.019 | 0.965 | 0.933 |
| amplitude_ratio_of_means | 0.965 | 0.946 | 0.718 | 0.619 |
| maximum_error | 0.343 | 0.163 | 0.259 | 0.336 |
| ks_mean_on_coarse_res_with_extremes | 0.139 | 0.176 | 0.285 | 0.373 |
| qqplot_mae | 0.772 | 0.775 | 0.838 | 1.141 |
| acf_mae | 0.049 | 0.065 | 0.139 | 0.167 |
| extremogram_mae | 0.026 | 0.021 | 0.032 | 0.052 |
Important: Right now we are only estimating the upper tail
extremogram. Currently we didn’t find a way to estimate the two tales at
the same time. We are using quant = .97
| Metric | cnn | naive | xgboost | lstm |
|---|---|---|---|---|
| diff_of_means | 0.287 | -0.319 | -0.337 | -0.412 |
| ratio_of_sd | 0.964 | 1.153 | 1.094 | 1.090 |
| amplitude_ratio_of_means | 0.934 | 0.962 | 0.922 | 0.817 |
| maximum_error | 0.312 | 0.112 | 0.216 | 0.299 |
| ks_mean_on_coarse_res_with_extremes | 0.194 | 0.304 | 0.294 | 0.349 |
| qqplot_mae | 0.848 | 1.031 | 0.983 | 1.203 |
| acf_mae | 0.045 | 0.086 | 0.097 | 0.122 |
| extremogram_mae | 0.042 | 0.100 | 0.020 | 0.019 |
Important: Right now we are only estimating the upper tail
extremogram. Currently we didn’t find a way to estimate the two tales at
the same time. We are using quant = .97
| Metric | naive | xgboost | lstm | cnn |
|---|---|---|---|---|
| diff_of_means | -0.134 | -0.170 | -0.241 | 0.295 |
| ratio_of_sd | 1.109 | 1.061 | 1.042 | 0.965 |
| amplitude_ratio_of_means | 0.961 | 0.884 | 0.769 | 0.944 |
| maximum_error | 0.114 | 0.224 | 0.325 | 0.326 |
| ks_mean_on_coarse_res_with_extremes | 0.287 | 0.299 | 0.401 | 0.265 |
| qqplot_mae | 0.631 | 0.541 | 0.703 | 0.876 |
| acf_mae | 0.068 | 0.096 | 0.122 | 0.034 |
| extremogram_mae | 0.034 | 0.025 | 0.037 | 0.028 |
Important: Right now we are only estimating the upper tail
extremogram. Currently we didn’t find a way to estimate the two tales at
the same time. We are using quant = .97
| Metric | naive | xgboost | lstm | cnn |
|---|---|---|---|---|
| diff_of_means | -0.165 | -0.198 | -0.266 | 0.269 |
| ratio_of_sd | 1.155 | 1.099 | 1.091 | 0.976 |
| amplitude_ratio_of_means | 0.963 | 0.901 | 0.799 | 0.940 |
| maximum_error | 0.108 | 0.215 | 0.316 | 0.322 |
| ks_mean_on_coarse_res_with_extremes | 0.276 | 0.297 | 0.373 | 0.233 |
| qqplot_mae | 0.878 | 0.724 | 0.778 | 0.787 |
| acf_mae | 0.085 | 0.106 | 0.127 | 0.046 |
| extremogram_mae | 0.053 | 0.020 | 0.022 | 0.028 |
Important: Right now we are only estimating the upper tail
extremogram. Currently we didn’t find a way to estimate the two tales at
the same time. We are using quant = .97